Ancient Siege Warfare

An extract from 'Siege Warfare in Classical Greece'

Classical Greek warfare was based on the punitive raid, designed to provoke the adversary into pitched battle; the accepted code of conduct obliged the two sides to meet in the ritualised clash of hoplite armies. Herodotus explains this in a speech that he puts into the mouth of the Persian Mardonius on the eve of Xerxes’ invasion of Greece (Hdt. 7.9). By and large, there was no question of capturing towns or enslaving enemy populations. Of course, many Greeks would have been familiar with besieging techniques: their cousins in Asia Minor had seen Persian siegecraft at first hand during the Ionian Revolt, and Greek mercenaries had served with the Great King’s army. But the resources of the average city-state would not have stretched to supporting the siege of a walled town.

Consequently, Greek armies lacked practice in this branch of warfare. This is illustrated by the Spartan attempt, around 525 BC, to overthrow the tyranny of Polycrates on the island of Samos. The Spartans initially gained a foothold on the town’s seaward wall, perhaps by escalade, but were ejected from the town by overwhelming forces. In the mêlée, two of their number rushed through the open gates, but were killed inside the town. The Spartans had reached the limit of their besieging skills and departed after 40 days (Hdt. 3.54–6). An event in 489 BC shows that contemporary Athenian siegecraft was just as rudimentary. In the aftermath of the Greek victory at Marathon, the Athenian general Miltiades tried to punish the island town of Paros for having aided the Persians. But, with the Parians safe within their walls, the only tactic available to the Athenians was to devastate the island, and after 26 days they gave up and left (Hdt. 6.133–5). The Spartan-led coalition outside Thebes in 479 BC found themselves in a similar predicament. The Persian invaders had just been defeated at nearby Plataea, ending their aspirations of Greek conquest, but Thebes was harbouring Persian sympathisers. Fortunately for the Spartans, the traitors gave themselves up after only 20 days of siege (Hdt. 9.86–7).

Athenian siegecraft

At some point, the Athenians acquired a reputation for siegecraft. The historian Thucydides says as much (1.102), although as an Athenian and a soldier he is perhaps a biased source. It is true that, following the battle of Plataea in 479 BC, the Spartans were unable to break into the Persian stockade, where the survivors were rallying, until the Athenian forces arrived (Hdt. 9.70). But when the Roman biographer Plutarch came to retell the story, it seems as though the Spartans were simply inexperienced in storming walls (Plut., Aristides 19). A comparison may be drawn with an incident from the battle at Mycale, allegedly fought on the same day as Plataea; during the rout of the Persian forces by the allied Greeks, it was the Athenian contingent that led the assault on the enemy stockade (Hdt. 9.102). As Garlan rightly observed, the supposed Athenian expertise was tested only against wooden palisades, not real fortifications.Another event from 479 BC points up this contrast. The Persian bridgehead on the European side of the Hellespont was based at Sestos, which remained in Persian hands even after Xerxes’ withdrawal from Greece. This strongly fortified town was strategically placed to challenge Athenian trade with the Black Sea region, so Xanthippus, the father of Pericles, led an Athenian fleet to capture it. The Persian governor Artayctes was unprepared for siege, so starvation quickly set in. Yet, despite this fact, the Athenians made little headway and complained to their officers, requesting to be taken home. It was only with the escape of Artayctes that the townsfolk were free to open their gates to the Athenians (Hdt. 9.114–121; Diod. Sic. 11.37.4–5). Similarly, the blockade of Persian-garrisoned Eion by Cimon, the son of Miltiades, was only brought to a conclusion when the Persian general Butes set fire to the place, preferring to perish than to be starved into submission (Plut., Cimon 7); the method of capture is not recorded by Thucydides (1.98), but a note by the Hadrianic traveller and writer Pausanias suggests that Cimon had diverted the town’s water supply (Paus. 8.8.9).

It is clear that the Athenians had developed no revolutionary besieging tactics. During the 470s and 460s BC, in the course of building their maritime empire under the guise of the Delian League, they often found it necessary to bring recalcitrant towns into line, but this was done not with an aggressive assault but by employing the costly method of blockade. Thasos provides a case in point. When the island revolted in around 465 BC, the ensuing Athenian siege dragged on into a third year before the Thasians surrendered; their walls were slighted, their navy was confiscated, and an annual tax was imposed (Thuc. 1.101). At Samos in 440 BC, Pericles is said to have erected blockading walls on three sides of the town (Thuc. 1.116), while Athenian ships patrolled the fourth side, which lay on the coast. When the ships briefly departed, the Samians took the opportunity to raid their oppressor’s fleet base and bring in supplies; but, on the return of the Athenian ships, the blockade was once more complete, and the Samians finally capitulated after nine months (Thuc. 1.117).

Blockading walls

Athens dealt similarly with Potidaea, which refused the unreasonable Athenian demands to dismantle its fortifications late in 432 BC. Sited on the westernmost finger of Chalcidice at the narrowest point, its walls ran from sea to sea, dividing the southern peninsula from the land mass to the north. The Persians had failed to storm the town during their retreat from Greece in 479 BC, but this was largely down to the incompetence of their commander (Hdt. 8.126–9). The Athenians adopted a different tack: two blockading walls were constructed, one to the north of the town and one to the south, completely barring the isthmus, and naval patrols watched both coasts (Thuc. 1.64). Unfortunately, the town proved surprisingly stubborn; in the second year of the siege, fresh forces from Athens made a vain attempt to storm the place using ‘machines’ (a word which Thucydides often uses to mean ladders), but their failure was compounded by an outbreak of plague and, after 40 days, they withdrew again (Thuc. 2.58). By this time, the Potidaeans had reportedly been reduced to cannibalism, and finally surrendered after a siege of over two years (Thuc. 2.70; Diod. Sic. 12.46.2–6).During the Peloponnesian War (431–404 BC), Athens used the technique of encirclement (periteichismos) several times, for example in 428 BC, when its erstwhile ally Mytilene on the island of Lesbos revolted. Here, the surrounding siege wall incorporated strongpoints for the Athenian garrison, but their blockade failed to prevent a Spartan agent from slipping into the town by way of a dry river bed. Fortunately for Athens, the Spartan plan to relieve Mytilene backfired when arms distributed among the townsfolk were used in a popular uprising. The town was handed over to the Athenians, who at first resolved to exterminate everyone, but were subsequently satisfied with the deaths of the 1,000 males who had taken part in the revolt (Thuc. 3.18, 25, 27–8, 36, 50). The Athenians were more ruthless at Melos, which they placed under siege in 416 BC for refusing to pay tribute. Different troop contingents vied with one another in building the encircling siege wall, and vigilance was increased after the townsfolk twice managed to bring in supplies through a weak sector. In the following year, when the town finally surrendered, all of the men were put to death, and the women and children were sold into slavery (Thuc. 5.114.1–2, 115.4, 116.2–3).

By that time, the periteichismos had become an Athenian hallmark. In 426 BC, the Acarnanian troops accompanying Demosthenes’ Athenian army at Leucas urged him to surround the town with walls, in order to speed the townsfolk’s surrender (Thuc. 3.94). Sieges occasionally took a different course, as at Mende, south of Potidaea, where internal squabbling between the townsfolk and the Peloponnesian garrison installed by Sparta gave the Athenians an ideal opportunity to burst into the town and subject it to wholesale plundering (423 BC); but the acropolis proved impregnable, so they resorted to constructing siege walls (Thuc. 4.130). Shortly afterwards, Mende’s neighbour, Scione, was surrounded with Athenian siege-works; but before the encirclement was complete, troops from Mende managed to break out and slip into Scione (Thuc. 4.131, 133). It did them little good, for they finally succumbed two years later and were put to death; the women and children were enslaved and the land given over to Athenian allies (Thuc. 5.32). As late as 409 BC, the technique was still in favour when the Athenians forced the surrender of Chalcedon by surrounding the town with a palisade (Diod. Sic. 13.66.1–3; Xen., Hell. 1.3.4–7); a similar strategy failed at Byzantium, until the town was betrayed from within (Diod. Sic. 13.66.3–67.7; Xen., Hell. 1.3.14–22).

The Athenian attack on the little island of Minoa in 428 BC demonstrates that, occasionally, a more direct approach was tried. The city-state of Megara, on the adjacent coast, had built a fort there, but the Athenian general Nicias captured it by landing ‘machines’ from the sea (Thuc. 3.51). The classicist Eric Marsden, best known for his work on ancient artillery, thought that these machines might have been ship-mounted siege towers. He was perhaps thinking of the transport vessel that the Athenians equipped with wooden towers for fighting in the harbour at Syracuse (Thuc. 7.25), but this could never have been used for an amphibious assault. It seems more likely that Nicias’ machines were simply assault ladders. The Athenians planned to use the island as a springboard for an attempt on the coastal town of Nisaea, which had been garrisoned by the Spartans’ Peloponnesian allies, but the attack, when it came, took the form of the familiar blockade.

The Megarians had built a pair of ‘long walls’ linking their town to Nisaea, in order to safeguard the route to the harbour there. However, in 424 BC, 600 Athenian troops crossed by night from Minoa and managed to infiltrate these walls, severing Nisaea’s link to Megara. Reinforced by 4,000 hoplites and gangs of stonemasons, they built a wall and ditch around the town, utilising materials salvaged from the suburbs and even incorporating entire buildings into their work; in two days, the periteichismos was complete and the Peloponnesian garrison surrendered (Thuc. 4.66–69).

Only at Syracuse did the Athenian strategy of periteichismos prove disastrous (see Plate B), but this can largely be blamed on poor leadership. In 415 BC, when Athens decided to extend its influence to Sicily by capturing this prosperous port city, an entire year was squandered in minor skirmishing, giving the Syracusans time to organise their defence and enlist Spartan assistance. Early in 414 BC, the Athenians took control of the Epipolae plateau overlooking the city, and built a fort at Syca as the pivotal point of the inevitable siege wall. By midsummer, a double wall extended south from the plateau to the harbour, despite Syracusan attempts to intercept it with palisades (Thuc. 6.97–101), but the Athenians delayed completion of the siege-works to the north. This fundamental blunder was quickly exploited by the Syracusans, under the leadership of the newly arrived Spartan general Gylippus. They intercepted the line of the siege wall with their own cross wall, which, adding insult to injury, was built with the stones already laid out for the Athenian wall (Thuc. 7.4–6). With one action, they had turned the tables on the Athenians, thwarting their blockade of the city.

The Oxford scholar G. B. Grundy memorably summed up the Athenians’ fame in siegecraft as ‘the reputation of the one-eyed among the blind’. But if their reputation was not based on conspicuous success, it perhaps arose from their ability to organise and finance the labour required to prosecute such an operation. Certainly, of the allied army besieging Miletus in 411 BC, it was the Athenian contingent that contemplated building siege walls (Thuc. 8.25). Of course, a blockade could be maintained without siege-works, but the Athenian predilection for constructing them suggests that real benefits were perceived. Quite apart from the protection of the besieging forces, and the concealment of their movements from the besieged, there were perhaps psychological factors involved, for the visual impact of a siege wall sent the message to the defenders that their plight was hopeless.